An accurate first-principles algorithm for predicting the 3D structure of 1-helical membrane proteins (MPs) from amino acid sequence would profoundly affect the course of biological research on MP function. The research proposed here is designed to bring us closer to that goal. The project is guided by our belief that the keys to successful prediction are to understand the physical principles of MP stability in lipid bilayers and the biological principles of MP assembly, especially the principles of transmembrane (TM) helix selection by the translocon complex (Sec61123 in eukaryotes and SecYEG in bacteria). The "Big Picture" goal of this project is to tighten the connection between the physical and biological principles as represented by hydrophobicity scales for predicting TM helices. We have discovered through molecular dynamics simulations of the SecYEG translocon from Methanococcus jannaschii, whose 3D structure is known, that the translocon is stabilized in its closed state by an intricate hydrogen-bond network that must be restructured during translocon opening initiated by signal sequences. Furthermore, we have found that this network is strongly perturbed by the well known prlA mutations of Escherichia coli that cause defects in protein secretion and helix insertion. Understanding the molecular basis for prlA-mutation defects will provide insights into the mechanism of translocon-mediated MP assembly. We thus propose to examine the effect of prlA mutations on the code used by E. coli translocons in vivo to select transmembrane helices during MP assembly. To connect prlA mutations to the selection code, we will develop in vivo TM-helix hydrophobicity scales for E. coli using single-span MPs to take advantage of the possibility of inserting TM helices along two different pathways: the SecA post-translational pathway or the co-translational signal recognition particle (SRP) pathway. This approach will clarify the relation between physical and biological principles. Our first steps in the development of a single-span MP system yielded unexpected and puzzling results. Even though single- span MPs are the most abundant MPs in all organisms, they have never been subjected to systematic study in E. coli. Preliminary results suggest the existence of previously unrecognized MP targeting signals that may involve the FtsH MP quality-control protease. These considerations lead to four specific aims: (1) Establish in vivo biological hydrophobicity scales for the insertion of single-span MPs along the SecA and SRP pathways. (2) Guided by molecular dynamics simulations, use the resulting scales of Aim 1 to examine the effects of prl mutants on SecYEG selection of TM helices. (3) To enhance genome-wide analyses of MPs, carry out a detailed bioinformatics analysis of E. coli single-span MPs in combination with systematic experimental studies to characterize and classify single-span MPs. (4) Characterize new targeting signals and a potential single- span MP insertion pathway that may involve the FtsH MP quality-control protease. PUBLIC HEALTH RELEVANCE: An accurate first-principles algorithm for predicting the three-dimensional structure of membrane proteins from amino acid sequence would profoundly affect research on membrane proteins, which are major targets for therapeutic drugs. The project is guided by the idea that the keys to successful prediction are to understand the physical principles of membrane protein stability and the biological principles of membrane protein assembly. Our goal is to tighten the connection between the physical and biological principles as a basis for prediction algorithms.